Mutations in the POLE gene encoding DNA polymerase e (Pole) have been linked to the development of colorectal and endometrial cancers with extremely high levels of hypermutation. The mechanism through which Pole variants cause hypermutability remains puzzling. Because amino acid residues in the exonuclease domain are preferentially affected, POLE mutations were originally thought to act by inactivating exonucleolytic proofreading. However, subsequent research by our laboratory challenged this concept. Mutator effects of cancer-associated variants in model systems greatly exceed the effect of proofreading deficiency (in the case of the most recurrent variant, P286R - by two orders of magnitude). PoleP286R mice were also found to be much more cancer-prone than proofreading-deficient Pole mice. Thus, factors other than loss of exonuclease must be responsible for the elevated mutagenesis and cancer risk. The goal of this application is to decipher the mechanisms that drive genomic instability in POLE-mutant tumors. Preliminary studies led us to propose that, for many Pole exonuclease domain variants, the main consequence is not merely the loss of exonuclease but a dramatically increased DNA polymerase activity, which affects the overall contribution of Pole to replication and interferes with the function of other mutation avoidance mechanisms. This is a highly innovative hypothesis, since it defines POLE variants in tumors as gain-of-function variants rather than loss-of-function as was previously thought. We will test this hypothesis by pursuing two Specific Aims. In Aim 1, we will use purified four-subunit Pole to characterize the biochemical effects of cancer-associated mutations. In Aim 2, we will define the mechanisms of the mutator effects of Pole variants in proliferating cells. These studies will help understand the etiology of a significant proportion of colorectal and endometrial cancers and may pinpoint new targets for the development of personalized therapies for POLE mutation carriers. The biochemical and cell- based assays we will establish in the course of this work will also be instrumental for assessing the functional significance of multiple poorly understood POLE variants detected in tumors.